Active learning in P-X phase space
Steps to run the files listed here:
1. Run the Active-learning-sh file in the cluster. 
2. Active-learning-sh file will run GP_mixtures.py file which will take data from training.csv and perform GP fit, and give out the accuracy and the next points to sample
3. The Shell script will add the next sample to training.csv, and step 2 will again be repeated unless the accuracy criteria gets satisfied.
4. Also, GP_mixtures will continue to add performance data to a mean.csv files like MRE, Correlation coefficent, Maximum GP error using the ground truth data, which is complete.csv
5. After the learning is finished, one can plots the final adsorption and error heat maps using the plotter data files from the Adsorption isotherm and error heat map folder.

PS: The process for P-X-T is the same expect the prior and test set distribution is different.